Cell-to-cell transmission of information in the CNS is thought to occur at individual synapses, isolated from their neighbors. However, phenomena such as transmitter spillover and pooling between synapses have suggested that point-to-point synaptic communication between neurons is not the only method of information transfer in the CNS. Whether spillover and pooling of transmitter occurs depends on several variables including the amount of transmitter released per synapse per presynaptic action potential. If clearance mechanisms are overwhelm by release, transmitter will spread away from synapses. At some synapses, manipulations that increase the probability of vesicular release result in the release of multiple vesicles from individual synapses following single presynaptic action potentials. This large bolus of transmitter can overwhelm local transporters and result in spillover. At other synapses, however, multivesicular release has not been found regardless of the release probability suggesting that there is a mechanism that prevents release of more than a single vesicle. Transmitter spillover and pooling are at least in part responsible for activation of extrasynaptic receptors including metabotropic receptors which in turn are required for several types of short and long term plasticity. The main objectives of this proposal are to determine the extent to which plasticity depends on transmitter spillover and the degree to which spillover depends on multivesicular release. This will be accomplished using a combination of electrophysiological and optical techniques in the study of three different synapses each with distinct characteristics. The plan is to use the attributes of each of these synapses to probe, first, the generality of multivesicular release and, second, the multiplicity of its consequences. All CNS functions including sensory processing, motor behavior, consciousness, and memory depend on intact synaptic transmission. Knowledge of the fundamental molecular and cellular properties of transmission is required for the development of rational interventions in pathological states.